Event-driven Architecture
4 likes•2,836 views
The document introduces event-driven architecture (EDA) as a flexible approach for designing complex software systems, emphasizing its ease of modeling in alignment with organizational communication structures, as stated by Conway's Law. Key benefits of EDA include scalability, ease of testing, and fault tolerance, making it well-suited for environments adopting microservices. The document also highlights a sample retail system implementation using EDA and provides resources for further reading.
Event-driven Architecture
- 4. A problem with ‘Monolith First’ A product of our agile approach to our projects…smaller features finished earlier. The goal is to deploy more oCen…this is hard. One soluDon is to have smaller applicaDons.
- 6. SOA Orders Invoicing Inventory Logging Profiles Warehouse
- 7. SOA Orders Invoicing Inventory Logging Profiles Warehouse Tightly Coupled
- 8. SOA with Enterprise Service Bus Orders Invoicing Inventory Logging Profiles Warehouse STILL Tightly Coupled Enterprise Service Bus
- 9. Event Driven Architecture • An Event is represents something that happens in a domain • Customer Submits Order • Customer Billed • Payment Received • Order Ready for Shipment • Order Shipped • While Events and their payload are designed at the enterprise system level, their implementaDons are leC to the specific subsystems.
- 10. Event Driven Architecture • No point-to-point integraDons • Loosely coupled, highly scalable systems • Loosely coupled, TEAMS • Easier to test • Easier to change • Topology agnosDc
- 11. Loosely Coupled Orders Invoicing Inventory Logging Profiles ReporDng Event Bus
- 12. Easy to Integrate New Systems Orders Invoicing Inventory Logging Profiles ReporDng Event Bus New System
- 13. Easy to Test
- 14. Inventory Inventory Inventory Scales Orders Invoicing Inventory Logging Profiles ReporDng Event Bus
- 15. The basic thesis of this article is that organizations which design systems (in the broad sense used here) are constrained to produce designs which are copies of the communication structures of these organizations. --- Melvin Conway
- 16. Easily Modeled as a finite state diagram
- 17. Other benefits • Ease of 3rd party integraDon • ExisDng systems can be ‘wrapped’ • Can be deliberate about scaling • Fault tolerant • Event messages can be logged and ‘replayed’ • Can test subsystems in isolaDon
- 18. A Retail System Sales Logger Invoicing Shipping Warehouse Rabbit MQ Exchange
- 19. A Retail System – PlaLorm AgnosMc Sales Logger Invoicing Shipping Warehouse Rabbit MQ Exchange
- 20. Order Accepted Event is generated Sales Logger Invoicing Shipping Warehouse Rabbit MQ Exchange
- 21. Customer Billed Event Sales Logger Invoicing Shipping Warehouse Rabbit MQ Exchange
- 22. Order Ready for Shipment Event Sales Logger Invoicing Shipping Warehouse Rabbit MQ Exchange
- 23. Order Shipped Sales Logger Invoicing Shipping Warehouse Rabbit MQ Exchange
- 24. Logger Consumes EVERY Event Sales Logger Invoicing Shipping Warehouse Rabbit MQ Exchange
- 25. Logger Consumes EVERY Event
- 27. For Further Reading • THIS presentaDon (and code) on GitHub hXps://github.com/shawnewallace/choreographed_process • Stephan Norberg EDA hXp://www.infoq.com/presentaDons/Domain-Event-Driven-Architecture • “Programming Without a Call Stack” hXp://www.enterpriseintegraDonpaXerns.com/docs/EDA.pdf • “Event-Driven Architecture Overview” • hXp://www.omg.org/soa/Uploaded%20Docs/EDA/bda2-2-06cc.pdf
- 28. Shawn Wallace Work: [email protected] Twitter: @ShawnWallace Blog: blog.shawnewallace.com http://www.about.me/shawnwallace